Authenticate with a seeded operator key; fold escalation and switches into TerdutTeam

Credentials: the TerdutServer controller generates <name>-operator-key in
the server's own namespace (owned by it) and hands it to the pods as
TERDUT_OPERATOR_KEY; the server creates its instance-scoped account from it
at every start. A replaced Secret rolls the pods. The bootstrap handshake,
the checkpoint Secret, per-team service accounts and credentials Secrets,
BootstrapStateLost and credentials.deletionPolicy are gone.

CRDs: TerdutServer, TerdutTeam and TerdutAlertSource. TerdutEscalationRule
and TerdutDeadmanSwitch become spec.escalation and spec.deadmanSwitches[]
on the team (matched by name, extras removed); team invites are removed.
A team is created under the identity <namespace>/<name> (external_id), so a
retry, a lost status or a deleted team heal by repeating the same call, and
a display name owned by another team is TeamNameTaken instead of an
adoption. The server resolves escalation usernames (UnknownUser condition).
OIDC claim names and trustEmail are spec fields.

Fixes: query values are URL-escaped; every delete treats 404 as success;
deleting a team no longer depends on allowedTeams consent; a switch or
integration deleted on the server is recreated; unnamed switches take the
CR's name.

Cleanup: scaffold e2e test, AGENTS.md, devcontainer, unused config/ pieces
and Client.Version() removed; DESIGN.md, README, ROADMAP and the demo
(run-demo.sh, manifests) rewritten for the new design. Secret RBAC stays
cluster-wide, now stated in DESIGN.md section 9.

Claude-Session: https://claude.ai/code/session_016mBLURvJoMuUEr9cB2RpUN
This commit is contained in:
Niklas Ye
2026-10-09 14:56:22 +02:00
parent b0a431f2a4
commit e1103f2b7d
92 changed files with 2082 additions and 7561 deletions
+22 -23
View File
@@ -3,6 +3,7 @@ package controller
import (
"context"
"fmt"
"maps"
"strings"
appsv1 "k8s.io/api/apps/v1"
@@ -31,7 +32,7 @@ func labelsFor(srv *terdutv1alpha1.TerdutServer) map[string]string {
// object (not just the desired one) so the caller can check
// status.readyReplicas.
func (r *TerdutServerReconciler) reconcileDeployment(
ctx context.Context, srv *terdutv1alpha1.TerdutServer, dbEnv []corev1.EnvVar,
ctx context.Context, srv *terdutv1alpha1.TerdutServer, dbEnv []corev1.EnvVar, keyHash string,
) (*appsv1.Deployment, error) {
deploy := &appsv1.Deployment{ObjectMeta: metav1.ObjectMeta{Name: srv.Name, Namespace: srv.Namespace}}
_, err := controllerutil.CreateOrUpdate(ctx, r.Client, deploy, func() error {
@@ -60,13 +61,8 @@ func (r *TerdutServerReconciler) reconcileDeployment(
deploy.Spec.Strategy = appsv1.DeploymentStrategy{Type: appsv1.RollingUpdateDeploymentStrategyType}
pod := srv.Spec.Pod
deploy.Spec.Template = corev1.PodTemplateSpec{
// pod.Annotations is assigned directly, not merged -- nothing
// else sets pod-template annotations today. If a future change
// needs the controller to set one of its own (e.g. a
// Prometheus-scrape annotation), this needs to become a real
// map merge with a stated precedence rather than silently
// clobbering one side.
ObjectMeta: metav1.ObjectMeta{Labels: labels, Annotations: pod.Annotations},
// The controller's own annotation wins over a same-named user one.
ObjectMeta: metav1.ObjectMeta{Labels: labels, Annotations: podAnnotations(pod.Annotations, keyHash)},
Spec: corev1.PodSpec{
EnableServiceLinks: new(false),
NodeSelector: pod.NodeSelector,
@@ -127,6 +123,17 @@ func (r *TerdutServerReconciler) reconcileService(ctx context.Context, srv *terd
return err
}
// operatorKeyHashAnnotation records which operator key the pods were started
// with, so replacing the key's Secret rolls them.
const operatorKeyHashAnnotation = "terdut.ryuvia.com/operator-key-hash"
func podAnnotations(user map[string]string, keyHash string) map[string]string {
out := make(map[string]string, len(user)+1)
maps.Copy(out, user)
out[operatorKeyHashAnnotation] = keyHash
return out
}
func servicePort(srv *terdutv1alpha1.TerdutServer) int32 {
if srv.Spec.Networking.ServicePort == 0 {
return 8080
@@ -194,13 +201,6 @@ func buildEnv(srv *terdutv1alpha1.TerdutServer, dbEnv []corev1.EnvVar) []corev1.
corev1.EnvVar{Name: "TERDUT_ARCHIVE_AFTER", Value: sweeper.ArchiveAfter},
)
deadman := srv.Spec.Deadman
env = append(env,
corev1.EnvVar{Name: "TERDUT_DEADMAN_MATCHERS", Value: deadman.Matchers},
corev1.EnvVar{Name: "TERDUT_DEADMAN_TIMEOUT", Value: deadman.Timeout},
corev1.EnvVar{Name: "TERDUT_DEADMAN_SEVERITY", Value: deadman.Severity},
)
notify := srv.Spec.Notify
if notify.NtfyURL != "" {
env = append(env,
@@ -230,6 +230,9 @@ func buildEnv(srv *terdutv1alpha1.TerdutServer, dbEnv []corev1.EnvVar) []corev1.
// here on" -- which is unconditionally true for anything this
// operator creates).
corev1.EnvVar{Name: "TERDUT_OPERATOR_MODE", Value: "true"},
// The credential this operator calls the API with; the server
// creates or re-keys its instance-scoped account from it at start.
corev1.EnvVar{Name: "TERDUT_OPERATOR_KEY", ValueFrom: secretEnvSource(operatorKeyRef(srv))},
)
if oidc := srv.Spec.OIDC; oidc.Enabled {
@@ -238,14 +241,10 @@ func buildEnv(srv *terdutv1alpha1.TerdutServer, dbEnv []corev1.EnvVar) []corev1.
corev1.EnvVar{Name: "TERDUT_OIDC_CLIENT_ID", Value: oidc.ClientID},
corev1.EnvVar{Name: "TERDUT_OIDC_NAME", Value: oidc.Name},
corev1.EnvVar{Name: "TERDUT_OIDC_SCOPES", Value: oidc.Scopes},
// terdut-server's own defaults for the claims/trust-email knobs
// the chart exposes but DESIGN.md's spec doesn't (§4.1's doc
// comment on OIDCSpec) -- not configurable here, not an
// oversight.
corev1.EnvVar{Name: "TERDUT_OIDC_USERNAME_CLAIM", Value: "preferred_username"},
corev1.EnvVar{Name: "TERDUT_OIDC_EMAIL_CLAIM", Value: "email"},
corev1.EnvVar{Name: "TERDUT_OIDC_GROUPS_CLAIM", Value: "groups"},
corev1.EnvVar{Name: "TERDUT_OIDC_TRUST_EMAIL", Value: "false"},
corev1.EnvVar{Name: "TERDUT_OIDC_USERNAME_CLAIM", Value: oidc.UsernameClaim},
corev1.EnvVar{Name: "TERDUT_OIDC_EMAIL_CLAIM", Value: oidc.EmailClaim},
corev1.EnvVar{Name: "TERDUT_OIDC_GROUPS_CLAIM", Value: oidc.GroupsClaim},
corev1.EnvVar{Name: "TERDUT_OIDC_TRUST_EMAIL", Value: boolString(oidc.TrustEmail)},
corev1.EnvVar{Name: "TERDUT_OIDC_SESSION_MAX_AGE", Value: oidc.SessionMaxAge},
)
if oidc.ClientSecretRef != nil {